linux-kernel - Re: [linux-keystone] [PATCH v2] drivers: cma: fix addressing on PAE machines

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <xa1tsj7l3agx.fsf@mina86.com>
Date:	Tue, 04 Dec 2012 20:04:30 +0100
From:	Michal Nazarewicz <mina86@...a86.com>
To:	Vitaly Andrianov <vitalya@...com>,
	Santosh Shilimkar <santosh.shilimkar@...com>
Cc:	m.szyprowski@...sung.com, kyungmin.park@...sung.com, arnd@...db.de,
	linux-kernel@...r.kernel.org, linux-keystone@...t.ti.com,
	Cyril Chemparathy <cyril@...com>
Subject: Re: [linux-keystone] [PATCH v2] drivers: cma: fix addressing on PAE machines

> On Tuesday 04 December 2012 06:37 PM, Michal Nazarewicz wrote:
>> They are all related to the very same issue, and what the whole patch
>> does is change the type used to store physical addresses from unsigned
>> long to phys_addr_t.  This is really a single change.

On Tue, Dec 04 2012, Santosh Shilimkar wrote:
> Thanks for clarification. 64 bit alignment fix and the allocation
> range checks can be two separate fixes and that is exactly what change
> log describes.  You have a last say though :-) No problem if you want
> to commit the patch as is.

I don't have strong feelings on this one, but I feel like it's really
a single change which manifests itself in a few ways.  If this is
confusing, maybe commit message could be improved, to something like:

------------- >8 -------------------------------------------------------
drivers: cma: represent physicall addresses as phys_addr_t

This commit changes the CMA early initialisation code to use phys_addr_t
for representing physical addresses instead of unsigned long.

Without this change, among other things, dma_declare_contiguous() simply
discards any memory regions whose address is not represtible as unsigned
long.

This is a problem on 32-bit PAE machines where unsigned long is 32-bit
but physical address space is larger.
------------- 8< -------------------------------------------------------

Vitaly, if you could resend with that description, it would be awesome,
and sorry for so much trouble in what appears to be a trivial patch. :P

-- 
Best regards,                                         _     _
.o. | Liege of Serenely Enlightened Majesty of      o' \,=./ `o
..o | Computer Science,  Michał “mina86” Nazarewicz    (o o)
ooo +----<email/xmpp: mpn@...gle.com>--------------ooO--(_)--Ooo--

Content of type "application/pgp-signature" skipped